One of the central paradigms of ecology is that only about 10% of organic carbon production of one trophic level is incorporated into new biomass of organisms of the next trophic level. Many of energy-yielding compounds of carbon are designated as 'essential', because they cannot be synthesized de novo by consumers and must be obtained with food, while they play important structural and regulatory functions. The question arises: are the essential compounds transferred through trophic chains with the same efficiency as bulk carbon? To answer this question, we measured gross primary production of phytoplankton and secondary production of zooplankton and content of organic carbon and essential polyunsaturated fatty acids of ω-3 family with 18-22 carbon atoms (PUFA) in the biomass of phytoplankton and zooplankton in a small eutrophic reservoir during two summers. Transfer efficiency between the two trophic levels, phytoplankton (producers) and zooplankton (consumers), was calculated as ratio of the primary production versus the secondary (zooplankton) production for both carbon and PUFA. We found that the essential PUFA were transferred from the producers to the primary consumers with about twice higher efficiency than bulk carbon. In contrast, polyunsaturated fatty acids with 16 carbon atoms, which are synthesized exclusively by phytoplankton, but are not essential for animals, had significantly lower transfer efficiency than both bulk carbon, and essential PUFA. Thus, the trophic pyramid concept, which implicitly implies that all the energy-yielding compounds of carbon are transferred from one trophic level to the next with the same efficiency of about on average 10%, should be specified for different carbon compounds.

a b s t r a c tDuring two sampling seasons we analyzed on weekly basis fatty acid (FA) composition of seston fraction o 130 mm and zooplankton fraction 4130 mm, and compared them using a multivariate canonical correlation analysis (CCA). Besides, we evaluated a possible impact of water temperature and inorganic nutrients on FA composition of the seston and the zooplankton.In spite of significant differences in percentages of several individual FAs, we found very strong canonical correlation (cross-correlation, 1-week lag) between FA composition of the seston and the zooplankton. The most important factor, providing the overall canonical cross-correlation between FA profiles of the seston and the zooplankton fractions was eicosapentaenoic acid (20:5o3, EPA). FA composition of the zooplankton fraction had comparatively poor correlations with taxonomic composition of the zooplankton. Thus, seasonal variations of FA composition of the zooplankton were determined primarily by seasonal changes in FA composition of the seston, rather than by taxonomic differences of FA profiles between rotifers, cyclopoids and cladocerans. FA composition of the seston was strongly affected by its taxonomic composition, namely by that of phytoplankton. According to CCA, the highest factor loadings pertained to diatoms interacting with their marker acids, including EPA, and cyanobacteria and greens, interacting with their marker acids. Ciliates and small rotifers composed considerable and sometimes major part of the seston biomass, but according to CCA their contributions to seasonal variations of the total FA profile of the seston were insignificant. This finding indirectly support the conclusion of the other authors, that the main source of FAs presented in ciliates and rotifers must be sought in algae and that they do not modify FA composition of food consumed, apart from repackaging it.Water temperature was the principal environmental parameter which drove the overall variations of FA composition. Factor loadings for the inorganic nutrients were comparatively negligible. The main contribution in the seasonal variation of FA composition of the seston was given by negative interaction between water temperature and percentage of EPA in the seston.

Summary We studied the fatty acid (FA) composition of six species of Cladocera and six species of Copepoda from five cold‐water lakes, situated in the tundra and/or in the mountains, and eight species of Cladocera and four species of Copepoda from eight warm‐water lakes (including one reservoir) in temperate regions. We asked whether the contrasting temperature would result primarily simply in changes in the percentages (i.e. percentage of total FAs) and absolute contents (quantities) of the long‐chain polyunsaturated fatty acids (PUFAs), eicosapentaenoic acid (20:5n‐3, EPA) and docosahexaenoic acid (22:6n‐3, DHA), or whether there are other FAs with various number of double bonds and/or chain lengths which could be responsible for a putative homeoviscous adaptation. We also aimed to reveal any consistent phylogenetic differences in FA percentages and contents between Cladocera and Copepoda, separable from any temperature effects. Both taxa in warm waters had greater percentages of 18:0, and lower percentages of 14:0 and 18:4n‐3, than in cold waters, but there were no differences in percentages of DHA. In addition, Cladocera, besides the lower percentage of EPA, had higher percentages of 20:0 and 22:0 in warm waters. These patterns in the percentages of 14:0, 18:0, 18:4n‐3, 20:0 and 22:0 are in a good agreement with the hypothesis of homeoviscous adaptation. Thus, the role of EPA, and particularly DHA, as unique regulators of the homeoviscous adaptation of the zooplankton may have been overestimated. Overall, we confirmed the known differences between Cladocera and Copepoda, namely higher percentages of EPA in Cladocera and higher percentages of DHA in Copepoda. However, there was c. 50% overlap in the ranges of the percentage of EPA in Cladocera and Copepoda, while the ranges in the content of EPA per unit organic carbon in Cladocera and Copepoda overlapped completely. Differences in the percentages and content of DHA between Cladocera and Copepoda were statistically significant and invariant with temperature, and therefore are probably due to phylogenetic factors, rather than any temperature adaptation. Contrasting temperature was not associated with significant differences in the contents of EPA and DHA per unit of organic carbon within the taxa studied. If this remained the case in a warming climate, such warming would be unlikely to reduce the accumulation of these important PUFAs in the zooplankton, at least if species composition was unchanged. However, if there were shifts in the proportions of Cladocera and Copepoda in the zooplankton, for example fewer copepods as temperature rises, a decrease of the flux of PUFA in the ecosystem is plausible, taking into account the phylogenetic (and temperature invariant) differences in DHA between the two groups.